Open AccessA Method for Hand Dorsal Vein Recognition Based on Frangi and Morphological Operation
Author(s) -
Xiaoling Xie,
Xiaorui Xu,
Junxi Zheng,
Xiaocui Sun
Publication year - 2025
Publication title -
ieee access
Language(s) - English
Resource type - Magazines
SCImago Journal Rank - 0.587
H-Index - 127
eISSN - 2169-3536
DOI - 10.1109/access.2025.3574188
Subject(s) - aerospace , bioengineering , communication, networking and broadcast technologies , components, circuits, devices and systems , computing and processing , engineered materials, dielectrics and plasmas , engineering profession , fields, waves and electromagnetics , general topics for engineers , geoscience , nuclear engineering , photonics and electrooptics , power, energy and industry applications , robotics and control systems , signal processing and analysis , transportation
Aiming at the problem of image recognition of the hand dorsal vein, a method for hand dorsal vein recognition based on Frangi and morphological operation (Frangi-M) is proposed. Frangi-M combines the Frangi algorithm and morphological operation. Firstly, the Frangi algorithm is used to enhance the vascular structure of the hand dorsal vein by using the difference of the Hessian matrix eigenvalues of the points in the vascular part and the non-vascular part of the image, and then morphological operations and median filtering are used to eliminate image spots and smooth the image. Frangi-M was compared to the adaptive threshold method and Butterworth high-pass filter method by using qualitative and quantitative evaluation methods. Both qualitative and quantitative evaluation results show that Frangi-M has the clearest and most complete identification of the hand dorsal vein. In the quantitative evaluation, the precision, recall, and mAP@0.5 of the YOLOv5 hand dorsal vein recognition model generated by Frangi-M are 0.90, 0.86, and 0.90, respectively, while those of the adaptive threshold method are 0.87, 0.82, and 0.89, respectively, and those of the Butterworth high-pass filter method are 0.85, 0.82, and 0.85, respectively.